云计算平台中多虚拟机内存协同优化策略研究  

Memory Cooperation Optimization Strategies of Multiple Virtual Machines in Cloud Computing Environment

作　　者：张伟哲[1] 张宏莉[1] 张迪[1] 程涛[1] 

机构地区：[1]哈尔滨工业大学计算机科学与技术学院,哈尔滨150001

出　　处：《计算机学报》2011年第12期2265-2277,共13页Chinese Journal of Computers

基　　金：国家自然科学基金(61173145);国家"九七三"重点基础研究发展规划项目基金(2011CB302605);国家"八六三"高技术研究发展计划项目基金(2010AA012504;2011AA010705)~~

摘　　要：虚拟化技术为云计算基础设施资源的动态部署、安全隔离提供了重要保证.从过度占用内存的虚拟机中回收内存,提供给内存紧缺的虚拟机使用,优化多虚拟机间的内存分布是内存虚拟化中的挑战性问题.文中引入了自发调节和全局调节协作的多虚拟机内存管理架构.通过定义内存资源充裕状态和内存资源紧缺状态,提出了自发调节和全局调节之间协作的算法.处于内存资源充裕状态时,各虚拟机可采用自发调节策略平衡内存资源.处于内存资源紧缺状态时,基于空闲内存价格的全局调节策略平衡内存资源.实验结果表明,针对计算密集型与存储密集型负载,该框架与算法均能很好地提高服务能力,同时具有较高的可扩展性与较低的性能惩罚.The virtualization technology has provided the important guarantee for the dynamic deployment and security isolation of the infrastructure as a service（IaaS） in the cloud computing.It is a challenging problem to optimize,across time,the distribution of RAM among a maximal set of virtual machines by reclaiming memory from ones that have an excess of memory and providing it to others that need more memory.This paper proposes a memory optimization scheme for multiple virtual machines with the cooperation of spontaneous adjustment and global adjustment algorithms.In the memory resources abundant status and scarce status,various virtual machines may adopt the spontaneous adjustment strategy and the overall adjustment strategy to balance memory resources,respectively.The experimental results indicate that in view of the computation intensity and the memory intensity load,this frame and the algorithm can sharpen the serviceability well,simultaneously have the high extendibility and the low performance penalty.

关 键 词：云计算 虚拟化 内存协同优化 自发调节算法 全局调节算法 

分 类 号：TP314[自动化与计算机技术—计算机软件与理论]